sonianuj287 commented on issue #62500: URL: https://github.com/apache/airflow/issues/62500#issuecomment-4043503991
After going through the discussion and experimenting with the Breeze workflow locally, one idea that might help keep the system maintainable is to treat Breeze commands as the canonical interface and derive the agent “skills” from them rather than defining skills independently. In other words, the skill layer could act as a thin machine-readable wrapper around Breeze commands. This would ensure that when Breeze evolves (new commands, flags, or workflows), the skills remain synchronized automatically instead of drifting over time. A possible approach could be: • introspecting Breeze CLI commands and metadata • generating or validating skill definitions from that metadata • enforcing this through a pre-commit check or validation step This would keep Breeze as the single source of truth while still giving AI tools a structured interface for common contributor workflows (running static checks, tests, starting services, etc.). One question I had while exploring this: should the skills system aim to cover *all* Breeze commands, or only a curated subset representing typical contributor workflows? I’d be happy to help prototype this direction if it aligns with the intended design. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
